Database-friendly random projections
PODS '01 Proceedings of the twentieth ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems
Privacy-preserving Distributed Clustering using Generative Models
ICDM '03 Proceedings of the Third IEEE International Conference on Data Mining
Very sparse random projections
Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining
International Journal of Computer Applications in Technology
Improving random projections using marginal information
COLT'06 Proceedings of the 19th annual conference on Learning Theory
Hi-index | 0.00 |
Privacy is the most important apprehension in many data mining applications. In this paper a new technique called Cryptic Random Projection, solves the re-identification quandary (which is found in the conventional random projections). Here this encryption based random projection assigns secret keys to the positions of random matrix elements and not to the random numbers. We have addressed two kinds of random sequences for generating the random sequences called determinist and indeterminist random sequences and encrypted it in a new way so that the original data cannot be re-identified. We have also optimized the privacy level which toughens the re-identification of original data without compromising the processing speed and data utility. We hope the projected solution will tarmac way for investigation track and toil well according to the evaluation metrics including hiding effects, data utility, and time performance.